造粒
过程分析技术
偏最小二乘回归
颗粒(地质)
关键质量属性
近红外光谱
材料科学
色谱法
粒径
硬脂酸镁
校准
分析化学(期刊)
化学
剂型
化学工程
数学
复合材料
量子力学
物理化学
工程类
统计
生物过程
物理
作者
Keita Koyanagi,Akinori Ueno,Tetsuo Sasaki,Makoto Otsuka
摘要
To produce high-quality pharmaceuticals, a real-time monitoring method for the high-shear wet granulation process (HSWG) was developed based on near-infrared spectroscopy (NIRS). Samples consisting of lactose, potato starch, and hydroxypropyl cellulose were prepared using HSWG with varying amounts of purified water (80, 90, and 100 mL) and impeller speed (200, 400, and 600 rpm), which produces granules of different characteristics. Twelve batches of samples were used for the calibration and nine batches were used for validation. After drying, the median particle size (D50), tapped density (TD), and Hauser ratio (HR) were measured. The best calibration models to predict moisture content (MC), D50, TD, and HR were determined based on pretreated NIR spectra using partial least squares regression analysis (PLSR). The temporal changes in the pharmaceutical properties under different amounts of water added and stirring speed were monitored in real time using NIRS/PLSR. Because the most important critical quality attribute (CQA) in the process was MC, granule characteristics such as D50, TD, and HR were analyzed with respect to MC. They might be used as robust and simple monitoring methods based on MC to evaluate the pharmaceutical properties of HSWG granules.
科研通智能强力驱动
Strongly Powered by AbleSci AI